Macrophage activation and coronary atherosclerosis in systemic lupus erythematosus and rheumatoid arthritis

Authors


Abstract

Objective

Activation of macrophages may contribute to increased atherosclerosis and coronary artery disease in systemic lupus erythematosus (SLE) and rheumatoid arthritis (RA). Neopterin, a pteridine derivative, is a novel marker of monocyte and macrophage activation that is associated with atherosclerosis and cardiovascular risk in the general population. We examined the hypothesis that macrophage activation is associated with accelerated atherosclerosis in SLE and RA.

Methods

We compared serum neopterin concentrations, adjusted for age, race, sex, and serum creatinine concentration, in patients with SLE (n = 148) or RA (n = 166) and control subjects (n = 177). In patients with SLE or RA, serum neopterin concentrations were then tested for association (adjusted for age, race, sex, serum creatinine, and medication use) with measures of disease activity or damage, inflammatory markers and mediators, and coronary artery calcium measured by electron beam computed tomography.

Results

Neopterin concentrations were significantly higher in patients with SLE (median 8.0, interquartile range [IQR] 6.5–9.8 nmoles/liter) and RA (median 6.7, IQR 5.3–8.9 nmoles/liter) than controls (median 5.7, IQR 4.8–7.1 nmoles/liter), and were higher in SLE patients than in RA patients (all P < 0.001). In SLE, neopterin was significantly correlated with higher erythrocyte sedimentation rate (ESR; P = 0.001), tumor necrosis factor α (P < 0.001), monocyte chemoattractant protein 1 (P = 0.005), and homocysteine concentrations (P = 0.01), but in RA, it was only associated with ESR (P = 0.01). Neopterin was not associated with coronary calcium in either SLE (P = 0.65) or RA (P = 0.21).

Conclusion

Macrophage activation, reflected by increased serum neopterin concentrations, was increased in both SLE and RA. Neopterin was more robustly associated with atherogenic mediators of inflammation and homocysteine in SLE than in RA, but was not associated with coronary atherosclerosis in either disease.

INTRODUCTION

Morbidity and mortality due to coronary artery disease are increased in both systemic lupus erythematosus (SLE) (1) and rheumatoid arthritis (RA) (2). The mechanisms underlying these increases are unclear, but inflammation is thought to play a role. The involvement of macrophages in atherosclerosis associated with SLE and RA has not been studied extensively, but is important since monocytes and macrophages play a key role in the early pathogenesis of atherosclerosis by ingesting oxidized low-density lipoprotein, transforming into foam cells, and recruiting additional monocytes and macrophages to the vessel wall (3).

Neopterin, a pteridine derivative, is a marker of macrophage activation. It is formed by GTP cyclohydrolase I that catabolizes GTP into 7,8-dihydroneopterintriphosphate, which is further metabolized to neopterin (4). Interferon-γ (IFNγ) is the most important inducer of GTP cyclohydrolase I, and therefore of neopterin formation. Therefore, neopterin is considered a marker of Th1 immune activation (4). Atherosclerosis has been described as a Th1-mediated disease (3), and higher concentrations of neopterin are associated with increased cardiovascular risk and atherosclerosis in the general population (5–8).

Several small studies indicate that concentrations of neopterin are elevated, or correlated with disease activity, in patients with SLE (9–13) or RA (14–18), but the association between neopterin and quantitative measures of atherosclerosis such as coronary artery calcification or carotid artery intima-media thickness has not been tested in these diseases. Therefore, we examined the hypothesis that macrophage activation is associated with atherosclerosis in SLE and RA by assessing whether neopterin concentrations are higher in patients with SLE or RA compared to controls and associated with disease activity, inflammatory mediators of cardiovascular risk, and coronary calcium in patients with SLE or RA.

SUBJECTS AND METHODS

Study population.

We studied patients with SLE (n = 148) or RA (n = 166) and control subjects (n = 177). All of the subjects were age ≥18 years and patients fulfilled the American College of Rheumatology classification criteria for SLE (19) or RA (20). The subjects are participants in ongoing studies of cardiovascular risk in cohorts of patients with SLE or RA; details regarding recruitment and methodologic procedures have been described (1, 2). By design, there were 2 separate groups of control subjects without inflammatory disease that were frequency matched for age, race, and sex to patients with SLE or RA (1, 2). In the present study, we combined these groups to form a single control group to allow comparison of neopterin concentrations between the 3 study groups with statistical adjustment for demographic variables. The study was approved by the Vanderbilt University Institutional Review Board and subjects gave written informed consent.

Measurements.

Clinical information, laboratory data, current and past medication use, and Agatston coronary calcium scores were obtained as described previously (1, 2, 21, 22). Briefly, coronary calcium was measured by electron beam computed tomography (CT) scanning with an Imatron C-150 scanner except for 37 patients with SLE and 5 controls, in whom a 64-row mulitdetector CT (LightSpeed VCT, General Electric) was used. For logistic reasons, scans were not performed in 3 patients. Coronary calcium was quantified as described by Agatston et al (23) by a single reviewer (PR) blinded to the clinical status of the subjects. Clinical indices of disease activity and damage, including the Disease Activity Score in 28 joints (DAS28) (24), Systemic Lupus Erythematosus Disease Activity Index (SLEDAI) (25), and Systemic Lupus International Collaborating Clinics (SLICC) (26) scores, were measured. Medication use was classified according to whether or not patients were currently taking corticosteroids, methotrexate (MTX), or antimalarials (both SLE and RA), or anti–tumor necrosis factor (anti-TNF) agents (only RA). Erythrocyte sedimentation rate (ESR), plasma homocysteine, and serum C-reactive protein (CRP) concentrations were determined in the Vanderbilt University Hospital clinical laboratory. Before 2003, the laboratory did not use a high-sensitivity CRP assay, and low concentrations were reported as <3 mg/liter; in 40 patients with SLE and 40 patients with RA who had CRP concentrations <3 mg/liter, CRP concentrations were measured by multiplex enzyme-linked immunosorbent assay (ELISA; Lincoplex Multiplex Immunoassay Kit, Millipore). Serum concentrations of TNFα, interleukin-6 (IL-6), and monocyte chemotactic protein 1 (MCP-1) were measured by multiplex ELISA (Millipore). Serum neopterin was measured by ELISA (ALPCO Diagnostics).

Statistical analysis.

Descriptive statistics were calculated as the frequency and proportion, mean ± SD, or median with interquartile range, according to the distribution of the variables. Demographic and clinical factors were compared by disease group using a Kruskal-Wallis test or Wilcoxon rank sum test, or a Pearson's chi-square test, as appropriate.

Serum neopterin concentrations were compared between patients with SLE or RA and control subjects using the Kruskal-Wallis test or Wilcoxon rank sum test for comparisons. A multivariable linear regression model was used to adjust for age, sex, race, and also serum creatinine, as renal function can affect neopterin concentrations (27).

Drugs used for the treatment of SLE and RA affect inflammatory activity; therefore, we compared neopterin concentrations separately in patients with SLE or RA who were currently receiving corticosteroids, MTX, antimalarials (both SLE and RA), and anti-TNF agents (only RA), and those who were not. Each comparison was performed using multivariable linear regression, adjusting for age, race, sex, serum creatinine, and use of the other drugs. For example, in SLE, when neopterin concentrations were compared in patients taking or not taking corticosteroids, the model was adjusted for the use of MTX and antimalarials. In RA, the analysis for corticosteroids was adjusted for the use of MTX, antimalarials, and anti-TNF agents.

Neopterin concentrations were examined for correlation with indices of disease activity and damage (SLE: SLEDAI and SLICC, RA: DAS28), inflammatory mediators (TNFα, IL-6, MCP-1, ESR, and CRP level), homocysteine, and coronary calcium score using Spearman's correlation coefficient in patients with SLE or RA. The relationships were further tested with adjustment for age, sex, race, serum creatinine, current corticosteroid, MTX, and antimalarial use in SLE, and additionally for anti-TNF agent use in RA, using either multiple linear regression (inflammatory mediators, DAS28, and homocysteine as the outcome variables) or proportional odds logistic regression (28). The proportional odds model was applied both to skewed continuous variables by using ranks for cut points and also to skewed ordinal variables, including the SLEDAI, SLICC, and coronary calcium.

Values for concentrations of inflammatory mediators, neopterin, homocysteine, and creatinine were log10-transformed when used in multivariate regression analysis in order to normalize the distribution. Statistical analysis was performed using R, version 2.9.2 (online at www.r-project.org). All tests were 2-sided and a P value of less than 0.05 was considered significant.

RESULTS

The clinical characteristics of patients with SLE or RA and control subjects are shown in Table 1. As expected, patients with SLE were younger and more likely to be women than patients with RA. As we have reported previously, concentrations of inflammatory mediators were higher in patients with RA or SLE than control subjects (22, 29), and serum IL-6 concentrations were higher in RA patients than in SLE patients (21). Only 1.7% of controls, 2.7% of SLE patients, and 1.8% of RA patients had a serum creatinine concentration >1.5 mg/dl.

Table 1. Characteristics of the study population*
 SLE (n = 148)RA (n = 166)Controls (n = 177)P
  • *

    Values are the median (interquartile range) unless otherwise indicated. SLE = systemic lupus erythematosus; RA = rheumatoid arthritis; HDL = high-density lipoprotein; LDL = low-density lipoprotein; SLEDAI = Systemic Lupus Erythematosus Disease Activity Index; SLICC = Systemic Lupus International Collaborating Clinics; DAS28 = Disease Activity Score in 28 joints; TNFα = tumor necrosis factor α; IL-6 = interleukin-6; MCP-1 = monocyte chemotactic protein 1; CRP = C-reactive protein; ESR = erythrocyte sedimentation rate.

  • Continuous variables compared with the Kruskal-Wallis test and categorical variables compared with Pearson's chi-square test (across all 3 groups).

  • P < 0.05 by Wilcoxon rank sum test comparing SLE and RA and P < 0.05 by Wilcoxon rank sum test comparing SLE and controls.

  • §

    P < 0.05 by Wilcoxon rank sum test comparing SLE and controls or RA and controls.

  • P < 0.001 by proportional odds logistic regression model comparing SLE versus controls or RA versus controls, adjusted for age, race, and sex.

  • #

    Available in 109 of 148 patients with SLE, 163 of 166 patients with RA, and 172 of 177 controls.

  • **

    P < 0.05 by Wilcoxon rank sum test comparing SLE and RA.

Age, years40 (30–48)54 (45–63)§47 (39–55)< 0.001
Male sex, %9.531.325.4< 0.001
White race, %66.988.6§78.5< 0.001
Hypertension, %44.6§53.0§28.8< 0.001
Diabetes mellitus, %4.711.4§2.80.003
Current smoker, %22.323.5§13.60.04
HDL, mg/dl48 (36–56)43 (37–54)45 (38–46)0.35
LDL, mg/dl94 (76–126)111 (88–134)117 (95–138)< 0.001
Triglycerides, mg/dl99.5 (73.2–147.0)112.0 (80.0–158.0)§93.0 (65.0–127.0)0.003
Creatinine, mg/dl0.8 (0.7–0.9)0.8 (0.7–0.9)0.8 (0.7–0.9)0.81
Agatston score0 (0–0.7)2.7 (0–150.4)0 (0–0)< 0.001
SLEDAI4 (0–6)
SLICC1 (0–2)
DAS283.89 (2.66–4.90)
TNFα, pg/ml#4.8 (3.1–7.4)§5.6 (2.8–11.0)§2.7 (2.0–3.9)< 0.001
IL-6, pg/ml#5.5 (1.9–26.2)13.9 (4.4–43.6)§2.5 (1.0–10.0)< 0.001
MCP-1, pg/ml#188.2 (124.7–294.6)§201.8 (145.1–289.3)§144.1 (98.8–206.1)< 0.001
CRP level, mg/liter3.6 (0.8–7.0)**4.0 (1.2–10.8)
ESR, mm/hour20.0 (9.0–34.0)16.0 (7.0–36.0)
Homocysteine, μmoles/liter9.2 (7.7–11.2)10.2 (8.1–11.9)§8.1 (6.9–9.5)< 0.001
Neopterin, nmoles/liter8.0 (6.5–9.8)6.7 (5.3–8.9)§5.7 (4.8–7.1)< 0.001

Serum neopterin concentrations were significantly higher in patients with SLE or RA than in controls, and were higher in SLE than in RA (all P values <0.001) (Table 1 and Figure 1). These differences remained significant after adjusting for age, race, sex, and serum creatinine (all P values <0.001). Serum creatinine was significantly associated with higher neopterin values (P < 0.001) in this model.

Figure 1.

Distribution of neopterin concentrations in control subjects and patients with systemic lupus erythematosus (SLE) and rheumatoid arthritis (RA). In the box plot, the horizontal solid line and the upper and lower hinges of the box show the median and the interquartile range, respectively. The whiskers extend to 1.5 times the upper and lower interquartile range. Neopterin concentrations were significantly higher in patients with SLE or RA compared to controls, and in patients with SLE compared to those with RA (all P < 0.001 with or without adjustment).

Medications were associated with neopterin concentration (Table 2). Current antimalarial therapy in SLE and current MTX use in RA were associated with lower neopterin concentrations. Anti-TNF agents did not affect neopterin levels in RA (P = 0.18).

Table 2. Neopterin concentrations in patients with SLE and RA and current medications*
 Neopterin (SLE)Neopterin (RA)
NMedian (IQR) nmoles/literPNMedian (IQR) nmoles/literP
  • *

    SLE = systemic lupus erythematosus; RA = rheumatoid arthritis; IQR = interquartile range; anti-TNF = anti–tumor necrosis factor.

  • Adjusted for age, sex, race, serum creatinine, and the current use of other drugs.

Corticosteroids      
 No637.6 (5.9–9.8)0.21766.6 (5.4–8.6)0.94
 Yes858.5 (7.1–9.8) 907.0 (4.9–8.9) 
Methotrexate      
 No1348.0 (6.5–9.8)0.87487.6 (6.0–10.7)0.01
 Yes148.6 (7.1–11.7) 1186.4 (4.9–8.3) 
Antimalarials      
 No498.7 (7.3–12.2)0.031247.1 (5.5–9.3)0.12
 Yes997.7 (6.3–9.3) 425.8 (4.8–7.5) 
Anti-TNF agents      
 No1326.9 (5.3–9.2)0.18
 Yes346.2 (5.0–8.0) 

The correlations between neopterin and disease indices, inflammatory mediators, homocysteine, and coronary calcium score in patients with SLE and patients with RA are shown in Tables 3 and 4, respectively. In SLE, neopterin correlated significantly with TNFα (ρ = 0.42, adjusted P < 0.001), MCP-1 (ρ = 0.31, adjusted P = 0.005), ESR (ρ = 0.33, adjusted P = 0.001), and homocysteine (ρ = 0.20, adjusted P = 0.01), but in RA, neopterin correlated only with ESR (ρ = 0.21, adjusted P = 0.01). There were marginal correlations between neopterin and the SLICC in SLE (Table 3), and the DAS28 in RA (Table 4). The association between the SLICC score and neopterin in SLE was attenuated by statistical adjustment that included serum creatinine in the model. Similarly, the univariate correlation between homocysteine and neopterin in RA (ρ = 0.24, P = 0.002) was attenuated by adjusting for serum creatinine (P = 0.11). Neopterin was not associated with coronary calcium in either SLE (P = 0.65) or in RA (P = 0.21) in adjusted analyses. When we additionally adjusted the results from Tables 3 and 4 for disease activity (SLEDAI and DAS28, respectively) the statistical significance of association between neopterin and coronary calcification did not change (SLE: P = 0.70, RA: P = 0.22).

Table 3. Correlations between neopterin and disease indices, inflammatory mediators, and coronary calcium in systemic lupus erythematosus*
FactorSpearman's correlation coefficient (ρ)PAdjusted P
  • *

    SLEDAI = Systemic Lupus Erythematosus Disease Activity Index; SLICC = Systemic Lupus International Collaborating Clinics; TNFα = tumor necrosis factor α; IL-6 = interleukin-6; MCP-1 = monocyte chemotactic protein 1; CRP = C-reactive protein; ESR = erythrocyte sedimentation rate.

  • Adjusted for age, sex, race, and creatinine, and current corticosteroid, methotrexate, and antimalarial use with multivariable linear regression.

  • Used proportional odds logistic regression.

SLEDAI0.080.340.49
SLICC0.190.020.08
TNFα0.42< 0.001< 0.001
IL-60.220.020.22
MCP-10.310.0010.005
CRP level0.060.480.21
ESR0.33< 0.0010.001
Homocysteine0.200.010.01
Agatston score0.020.770.65
Table 4. Correlations between neopterin and disease indices, inflammatory mediators, and coronary calcium in rheumatoid arthritis*
FactorSpearman's correlation coefficient (ρ)PAdjusted P
  • *

    DAS28 = Disease Activity Score in 28 joints; TNFα = tumor necrosis factor α; IL-6 = interleukin-6; MCP-1 = monocyte chemotactic protein 1; CRP = C-reactive protein; ESR = erythrocyte sedimentation rate.

  • Adjusted for age, sex, race, and creatinine, and current corticosteroid, methotrexate, antimalarial, and anti-TNF agent use with multivariable linear regression.

  • Used proportional odds logistic regression.

DAS280.140.070.08
TNFα0.170.030.22
IL-60.040.630.75
MCP-10.120.130.24
CRP level0.130.100.07
ESR0.210.010.01
Homocysteine0.240.0020.11
Agatston score0.200.010.21

DISCUSSION

The major finding of this study is that concentrations of neopterin, a marker of macrophage activation and IFNγ activity, are increased in large cohorts of patients with SLE or RA, and have a more robust association with mediators of inflammation and homocysteine in SLE than in RA.

Previous studies of neopterin in SLE (9–13) and RA (14–17) have been limited largely to defining the relationship between neopterin and disease activity. Generally, they found that neopterin correlated with disease activity in both SLE and RA, and decreased with treatment. Therefore, neopterin has been considered as a candidate biomarker for monitoring disease activity. However, these studies most often used the urinary neopterin to creatinine ratio as a biomarker. This ratio adjusts neopterin excretion for variation in urine concentration rather than renal function (30, 31). We found that neopterin correlated marginally with the DAS28 in RA, but not with the SLEDAI or SLICC in SLE. The SLICC score measures damage in SLE and encompasses renal damage. The lack of association between neopterin and the SLICC score after adjustment that included serum creatinine was likely due to the fact that both were positively associated with serum creatinine concentrations. The influence of renal function as well as medications on neopterin concentrations suggests that its usefulness as a biomarker for disease activity is limited.

The lower concentrations of neopterin found in patients with SLE taking antimalarials and in those with RA taking MTX suggest that these drugs may inhibit monocyte and macrophage activation. Such associations are interesting, but should be interpreted cautiously, given the cross-sectional design of the study.

Higher neopterin concentrations in patients with SLE compared to those with RA have also been observed by others (9). This observation is counterintuitive, since current thinking is that SLE is a Th2-mediated disease (32), RA is a Th1-mediated (or more recently, Th17) disease (33), and neopterin is a marker of Th1 immune activation (4). However, several recent lines of evidence implicate IFNγ in the pathogenesis of SLE (34), and therefore our finding of higher neopterin concentrations in patients with SLE is concordant with those observations.

We have recently reported that TNFα and IL-6 were independently associated with coronary calcification in SLE and RA (22, 29, 35). The stronger association of neopterin with TNFα, IL-6, and MCP-1 (a mediator of monocyte activation) (36) in SLE than in RA, along with the higher neopterin concentrations in SLE than in RA, suggest that the activation pathways that drive neopterin production may differ in the 2 diseases, and that neopterin is associated more closely with atherogenic pathways in SLE than in RA.

Neopterin has been associated with atherosclerosis and coronary risk in the general population. Higher concentrations of neopterin were associated with increased carotid (5) and coronary atherosclerosis (6), and were of prognostic value in patients with acute coronary syndrome (7). These clinical observations suggested a role for monocyte and macrophage activation in the pathogenesis of atherosclerosis and its complications, and were concordant with histologic studies in which monocytes and macrophages were prominent in atherosclerotic lesions (37).

Although neopterin concentrations were correlated with some important atherogenic inflammatory mediators in SLE, they were not independently correlated with coronary calcium in either SLE or RA. It is possible that in studies finding an association between neopterin and increased cardiovascular risk (5–7) or carotid artery plaque (8) in the general population, this may have occurred indirectly. For example, if neopterin concentrations and inflammatory mediators were correlated, the observed association between neopterin and atherosclerosis could actually have been due to the relationship between atherosclerosis and inflammatory mediators that were not measured. Alternatively, in chronic inflammatory diseases such as RA and SLE, the increase in neopterin concentration could be driven by generalized inflammation, and this could act to obscure any smaller neopterin signal produced by macrophages associated with atherosclerosis. Finally, some studies suggest that neopterin may identify a patient group with increased cardiovascular risk that is not related to the amount of atherosclerosis present (38). We cannot exclude this possibility in our study.

Homocysteine is a strong predictor of cardiovascular risk (39), and others have also noted an association between homocysteine and neopterin concentrations in a range of populations (17, 40, 41), including a study of 33 patients with RA (17). The association between neopterin and homocysteine concentrations is thought to occur because immune activation results not only in higher neopterin concentrations but also in oxidative stress, consumption of folate, and impaired metabolism of homocysteine (42). Therefore, neopterin and homocysteine levels may both reflect immune activation. Also, because both neopterin (27) and homocysteine (43) are cleared by the kidney, their concentrations would both tend to track with renal function. Our observation that the association between neopterin and homocysteine was attenuated when creatinine was included in the model in RA, but not in SLE, is compatible with the idea that neopterin has a stronger association with inflammation in SLE than in RA.

Our study has several limitations. It was cross-sectional in design, and therefore causal inferences cannot be drawn. Also, we measured coronary calcium, a technique that provides an excellent measure of the atherosclerotic burden, but does not discriminate between stable and unstable lesions. The prognostic association between neopterin concentrations and acute coronary syndrome (7) suggests that additional studies will be required to determine whether the increased concentrations of neopterin in patients with SLE and RA could reflect a greater burden of unstable atheromatous plaque, or an increase in the processes by which stable plaque becomes unstable.

In conclusion, macrophage activation, reflected by increased serum neopterin concentrations, is increased in both SLE and RA compared to control subjects. Neopterin is more robustly associated with atherogenic mediators of inflammation and homocysteine in SLE than in RA, but is not associated with the severity of coronary atherosclerosis in either disease.

AUTHOR CONTRIBUTIONS

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Stein had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Rho, Oeser, Stein.

Acquisition of data. Rho, Solus, Raggi, Oeser, Stein.

Analysis and interpretation of data. Rho, Solus, Raggi, Gebretsadik, Shintani, Stein.

Ancillary